Configurations.md

+## Data Preparation
+Please refer to <code>data_prep_config.yml</code> for the following configurations.
+
+### Global Configuration
+| Setting| Default|Description|
+| ------ | ------ |------ |
+|    iso_code    |(various) | ISO code of the language |
+
+### Corpus Configuration
+| Setting| Default|Description|
+| ------ | ------ |------ |
+|    raw_data_file |(path_to_file) | Location of the corpus, please combine multiple files into a single file |
+|    raw_data_type |toolbox | Filetype of the corpus (supported: toolbox, todo: [flex, elan, odin, pangloss]) |
+
+### Toolbox Corpus Configuration
+| Setting| Default|Description|
+| ------ | ------ |------ |
+|    toolbox_transcript_tier|\\tph | Name of transcript tier in Toolbox file |
+|    toolbox_morpheme_tier|\\mph| Name of morpheme tier in Toolbox file |
+|    toolbox_gloss_tier|\\mgl| Name of gloss tier in Toolbox file |
+|    toolbox_pos_tier|\\ps| Name of part of the speech tier in Toolbox file |
+|    toolbox_translation_tier|\\eng| Name of English translation tier in Toolbox file |
+
+### XIGT Configuration
+| Setting| Default|Description|
+| ------ | ------ |------ |
+|    raw_to_xigt | true | Set to true to turn on the raw corpus to XIGT conversion |
+|    raw_to_xigt_path | (path_to_XIGT_file) | Location of to-save XIGT file, only applicable when raw_to_xigt=true |
+|xigt_path| (path_to_XIGT_directory)| Directory where stores all XIGT files for this language|
+|xigt_file_path| (path_to_XIGT_file) | Load existing XIGT file, only applicable when raw_to_xigt=false|
+
+### XIGT Enrichment Configuration
+| Setting| Default|Description|
+| ------ | ------ |------ |
+|    xigt_to_enriched_xigt| true | Set to true to turn on XIGT enrichment |
+|    enriched_xigt_file_path | (path_to_enrichedXIGT_file) | Path to store the enriched XIGT file |
+|split_enriched_xigt|true| Set to true to split the enriched XIGT into 10 training folds and 10 test folds|
+
+### Testsuites Configuration
+| Setting| Default|Description|
+| ------ | ------ |------ |
+|    create_testsuites| true | Set to true to create testsuits from enriched XIGT |
+| testsuites_directory_path | (path_to_testsuits_directory)| Directory to store all testsuits|
+
+### AGG Config Preparation
+| Setting| Default|Description|
+| ------ | ------ |------ |
+|prepare_agg_configs|true|Set to true to run AGG config file preparation|
+|agg_config_path|(path_to_agg_config_directory)| Directory to store AGG config files|
+| collect_tags | true | Set to true to collect POS tags set from enriched XIGT|
+|pos_tag_tier_names| "pos m" | POS tag tier names in enriched XIGT|
+
+## AGG Inferences Configuration
+Please refer to <code>agg_config.yml</code> for the following configurations.
+| Setting| Default|Description|
+| ------ | ------ |------ |
+|agg_config_path|(path_to_agg_config_directory)| Directory of the AGG config files|
+|compression_rounds|1|TODO|
+|output_dir|(path_to_output_directory)|Directory to output inferred choices file and skipped items|
+|graph|false|TODO|
+|hyphens|true|TODO|
+|precluster|None|TODO|
+|glosses|true|TODO|
+|compression|0.2|TODO|
+|lexitem_classes|true|TODO|
+|all_stems_occur_bare| true|TODO|
+|ignore_chars| None|TODO|
+|ungrammatical| '*!?'|TODO|
+|allomorphs| None|TODO|
+|boundaries| true|TODO|
+|allow_difference| 5|TODO|
+|escape_special_characters| true|TODO|
+|wordlist| ''|TODO|
+|infer_case| gram|TODO|
+
+## Evaluation Preparation
+Please refer to <code>eval_config.yml</code> for the following configurations.
+| Setting| Default|Description|
+| ------ | ------ |------ |
+| TODO||
\ No newline at end of file